RMS question (don't go mad one answer will do)

I have some experience in building engines from when I used to race saloon cars about 3 years ago. I have never taken a Porsche engine apart yet, put have seen various diagrams and I have an understanding of its design and the RMS. It is simply an oil seal which inner seals on the crankshaft and outer seals on the engine casting (like most other engines do), and when these start to fail they weep a small amount of oil and if not repaired the leak becomes worse, and that is about it!! I know it is annoying having a small pubble of oil in your garage where you park overnight, and you would expect more from a Porsche, but if you keep your oil topped up there is no real danger to your engine. My question is simply this, how can an RMS failure be attributted to an engine failure?? (I am, of course, assuming that everyone checks their oil regularly). Simon Hill

When we had RMS leak - we did not lose any oil at all. Engine failure as nothing to do with RMS leaks.

hmmm two questions in one. Yes in most engines rear main seals (rear bearing/flywheel oil seal) cause nothing but a minor leak and only after many thousands of miles. The difference here is that these seals are leaking at low miles. Porsche has tried I think three seals so far and to the best of my knowledge they have yet to cure the problem totally. New seals apparently are much better though.And new methods of fitting have helped. What you might be tempted to ask is why are apprently new good seals leaking in the first place. Given that older design engines only tend to leak when perished and when main bearings have started to wear. The seal is there to catch any oil passing through the main bearing and sling it back. In most engines this oil is absolutely minimal due to the tight tolerances involved. A rms failure as correctly pointed out cannot on it's own cause an engine failure. There is a school of thought that asks the question "given that all three seals have yet to cure the problem, why is there oil passing newish bearings in sufficent amounts to cause a leak in the first place" "is there a flaw in the fundamental design of the engine causing the rear main bearing to leak oil and eventaully fail" So no a rms seal cannot cause engine failure. But an out of round "possibly" rear main bearing can. Symtom rear main seal leaking. Cause o.o.r rear main? The M96 engine features a very offset flywheel in relation to the crankshaft main bearing and also a crankshaft "cage" design to house the crank and bearings , this is seprate but rigidly bolted to the engine case and misalignment or movement however small "could" cause problems. Porsche are likely to never let us know. As I say this is a school of thought there is no hard evidence to back this up.

interesting article on RMS in this months Total 911

Hello Laurence, you mentioned "leaking main bearings" do you mean the big end bearings? If you do, usually called " big end shells" - these are supposed to leak oil as they are feed by oil galleries which run inside the main crankshaft and through the shell. If you excessive wear in this bearing, the engine would probably sound like a diesel !! I don't think that acceptable big end bearing wear would lead to main seal failure for the simple reason that the amount of crankshaft movement required to cause a flexible rubber seal to leak would be a considerable amount - so much so the your engine would sound totally knackered. Trust me - have destroyed quite a number of race engines over the years. I have yet to actually see an engine which has had an RMS failure, and would very much like to see one if anyone out there knows of one which is available to see. My personal opinion, is that the fault is with the type of seal used, rather than engine design - but I will reserve judgement until I can actually assess an engine with this problem. Also - how is it possible to have an offset flywheel in relation to the crankshaft?

Erm no i mean't Main bearings (as in the ones that mount the crankshaft and allow it to rotate!) rather than "big end shells" as this could confuse you into thinking I was talking about big ends (con rod variety) I was in no way implying that the bearings did not pass (leak is incorrect I think) oil through them (by there very nature they do this in order to work) what I said was that this was minimal and was handled by the rms which acts to not only stop the oil but also to "sling" it back. Offset flywheel was the best way I could describe the way in which the flywheel is mounted in position to the crankshaft and it's main bearings. The M96 engine has an engine case. The crankshaft and connecting rod assemblies are mounted in a " bearing cage" which caries nearly all the the main bearing sadles and creates a ridged "box" into which the crankshaft is bolted into this in turn is bolted through the engine case, the remaining bearings being the flywheel and front. The best way to see this is in engine diagrams and drawings. The M96 engine is quite a departure from any other flat six design and was built to cost.Older M64 and previous air cooled designs feature a crankcase with machined in bearing saddles that are integral to the design and creat considerably more strength. I would beg to differ that the bearing would have to be completely shot or that the engine would exhibit any greater amount of noise. Any high mileage air cooled flat six stands a good chance of exhibiting the same symptoms. The bearings don't have to be shot to pieces to do this. It IS strange though that engines covering 20k -30k and less have shown these symptoms.One day dry next day it leaks ....why? has the seal suddenly become age hardened? Or is something else causing the seal to prematurely fail. I have an open mind on this.

You are, of course, correct about me getting my mains and big ends mixed up (if you pardon the expression!!) - I realised what I had written about 10 minutes ago - hence the rapid reply. Interesting point about the cage - I wasn't aware of this design, can I assume then that the block is cast seperately and then a "type of scaffolding" (for want of better word) is then inserted into the casting and bolted in, and it is this scaffolding that the crank main journals are machined into? Wierd?

Just thought of something! Is there a link between the type and style of driving that may cause early rms failure. Bear with me for a while - I will try and put into some kind of sensible text. The seal is positioined in an area between the flywheel and the engine, we all know that sitting in traffic with stop / start driving conditions will generate a lot of heat in the area. Could prolonged conditions like this lead to the premature perishing of the rubber? Is it possible that most of the rms failures are on cars that are mainly driven in these conditions, rather than the motorway driver who is driving at maintained speeds which aid good airflow and thus better all round cooling? I'm not saying that people are driving their cars wrong in traffic (that would be a bit silly as there is nothing anyone can do about traffic), but it may shed some light onto the real cause of the failure. (i.e. is it linked to high temperatures and high duration?) Only a thought?

But a very interesting one, none the less.

From what I have understood of the countless previous posts about the RMS, there is no statistical link between engine failures and RMS failures. The problems seem to be twofold 1) The cost of replacement(s) (in some cases up to 7-8, wth some seals requiring replacement with less than 1,000 miles from new) 2) Clutch contamination. This I believe is very rare, and can be left until the clutch naturally needs replacement anyway. Tiptronic cars apparently have less reported cases of RMS leaks, but are more costly to repair. But does the seal leaking signal possible future problems with the engine? Scouser (a multiple RMS sufferer) had this to say one year ago: The seal is not the problem. The problem is the engine block casing. It is due to a misalignment of the engine block casing which has become non concentric to the crankshaft. It is my understanding that the engine block casing is constructed of an amalgam of materials (mainly aluminium). The design allows for expansion due to thermal fluctuations in engine operating temperatures. However, the design has shown to lead to a warping of the engine block casing resulting with the opening bore of the crankshaft where the RMS fits to be non-concentric. Since the RMS is circular it cannot possibly be fitted correctly into a bore cavity that has become oval. The result is that the seal leaks by allowing oil to seep through it and into the gearbox bell housing and eventually onto the floor of one's garage. It does not matter how many seals are replaced, the result will always be the same: any replaced seal will eventually leak. If left unattended a failed RMS may lead to the collapse of the main bearing or the untimely wear of the engine due to crankshaft wobble and thus ultimately leading to catastrophic engine failure. It may also lead to clutch and/or flywheel failure due to oil contamination from the oil leaking into the gearbox bell housing. PAG have tried sevral seal types in an effort to fix this problem but so far they haven't worked'. However, as I mentioned, there is no hard evidence to link a leaking RMS with an engine failure. I believe that most of the M96 engine failures have been due to slipped liners and bearing failures...it would be very interesting to know if the bearing failures were always preceded by RMS failures........also RMS failures have also been linked with the Getrag gearbox failing. The point is this....not knowing if an RMS leak is a precursor to engine failures/gearbox failures, do you actively look for cars with RMS leaks that have previously been repaired (as they are less likely to fail more often, therefore less RMS replacement expense) or do you avoid them like the plague to avoid impending possible engine/gearbox death?

here is a picture of the crankshaft cage assembly being lowered into the engine case. Picture is from Porsche Website hope i've not broken any copy rules) please remove if neccessary.

Good point - well put! I had the same thought when looking to buy my car, bought one with no recorded rms problems and also got the extended warranty - so not really bothered either way. Interesting to read what scouser wrote, has the misalignment actually been measured by an independent engineer and confirmed? Is the "bird cage" which is bolted into the block made of a material which has the same thermal expansion properties as the aluminnium casing? You would have expected a company like Porsche to have explored these things prior to production bearing in mind that they are self-confessed "engineers". Or perhaps it is due to the fact that Porsches have become mass produced cars now - you know what mass production = larger manufacturing tollerances!!! Sorry if I am asking questions which have been asked before, I am relatively new to this forum and do not wish to open the proverbial "can of worms". I'm just interested in trying to solve anything which has some kind of mechanical mystery - that's all.

Laurence, just seen the picture of the cage, seems like a fairly heafty bit of kit and up to the job in hand. I guess we should remember when car manufacturers started using wet liners in aluminium blocks, everyone thought they were mad, expecting most of them to rattle out within a week or so - they're built like that now!! Does the new 997 model use the same engine design?

997 design is an updated version of same.Strictly from memory, the Porsche water cooled flat six engine uses a system which casts the liners into the bore. Has some special name which escapes me at the mo. But early boxsters suffered from porrous liners. That problem now fixed. The cage meerly holds the crankshaft and rods. Sadly in the pic you can't see the flywheel end. If you go to the Porsche website and go to the factory virtual tour you will see some more engine assembly pic. Gotta plug the website seeing as I borrowed a piccy[ ]

"The problem is ... due to a misalignment of the engine block casing which has become non concentric to the crankshaft. It is my understanding that the engine block casing is constructed of an amalgam of materials ....... however, the design has shown to lead to a warping of the engine block casing resulting with the opening bore of the crankshaft where the RMS fits to be non-concentric ...... PAG have tried several seal types in an effort to fix this problem but so far they haven't worked'. So how do 993 / GT3 & 996TT blocks differ then? Fundamental design and / or materials used? Mass production should produce a less variable product as there should be the volume to invest in better machine tooling.... Materials? We're talking Porsches here, not Ladas. The cost of different Aluminium alloys etc shouldn't matter. Is there a fundamental design flaw that racing / competition testing would have kept the cooking 986/996/987/997 block on the drawing board?

Another description by WP1206 22/8/04 on tis thread http://www.porscheclubgbforum.com/fb.asp?m=71682 'This may be of interest to you all. I talked to a very knowledgeable long time engine builder recently. Looking over a 993 engine in pieces in his workshop the subject of 996/Boxster engine design came up. He used the disassembled 993 engine to talk me through both the 993 and 996/Boxster engine designs. The 996/Boxster engine has some very nice design elements. The subject came around eventually to RMS and engine failures. I do not have the specialist knowledgeable to know if my memory or understanding is 100% accurate but wanted to post the result of the discussion to maybe help with your ongoing discussions as all knowledge is good. I can claim a mechanical engineering background but the following should be taken as non-official, unverified, blah blah blah. I will make the description as simple as possible whilst describing the relevant points. I'll start by describing the crankshaft arrangement in the engines pre 996/Boxster that were last seen in the 993. In these engines the crankshaft arrangement is something like this. a/ Starting at one end of the crankshaft you have the intermediate shaft drive. The intermediate shaft is driven directly by having a gear wheel on the crankshaft and a gear wheel on the intermediate shaft. b/ In the middle are the connecting rods to the pistons. c/ At the other end of the crankshaft is the final crankshaft bearing (I have not described any other bearings as not directly relevant) followed immediately by the rear main seal (RMS) followed by the dual mass flywheel outside the crankcase. For the 996/Boxter engine redesign a number of things where changed in this layout. First the intermediate shaft drive was changed from direct gear wheel to a chain and sprocket type of drive. Also the intermediate shaft drive was moved to the other end of the engine between the final crankshaft bearing and the rear main seal (RMS). So in summary the layout looks like this a/ Starting at one end of the crankshaft you have some stuff (no detail given here as not directly relevant). b/ In the middle are the connecting rods to the pistons. c/ Then the final crankshaft bearing (I have not described any other bearings as not directly relevant). d/ Then the intermediate shaft drive (sprocket and chain). e/ Then the main seal (RMS). f/ Finally the dual mass flywheel outside the crankcase. The possible fault with this design is the distance between the final crankshaft bearing and the dual mass flywheel. What happens is that the crankshaft bends under use probably exacerbated by the weight and therefore the rotational forces caused by the heavy dual mass flywheel. Once the crankshaft has bent it starts to wobble at the flywheel end and no longer revolves in true around its centre line. This could cause the following problems. a/ Rear main seal (RMS) to fail. As the crankshaft is bent is revolves out of centre and the seal is not designed to deal with this. Eventually the seal no longer forms a full seal to the crankshaft as it gets deformed in effect to have a slightly larger hole in the middle through which the oil leaks. b/ Failure of the intermediate shaft and therefore catastrophic failure of the engine. Because the crankshaft is no longer true this results in the chain to the intermediate shaft being stretched then loosened on every revolution. It’s a sort of jerking movement. It would eventually cause either the chain drive to fail or the intermediate shaft itself to bend with subsequent damage to bearings etc. c/ Vibrational forces at the rear of the crankshaft around the rear bearing. It’s unknown what this might cause' Another link to the RMS survey thread http://www.porscheclubgbforum.com/fb.asp?m=129297 The GT3 and TT have the same crankshaft, cylinder liners and cylinder blocks, all based on the GT1, and different to the M96 engine, that's why they do not suffer from RMS failures. .

"Materials? We're talking Porsches here, not Ladas. The cost of different Aluminium alloys etc shouldn't matter." [/quote] Don't tell me Ladas have RMS problems as well. What is the world coming to?

Very informative article in the latest issue of Total 911 on RMS.

The crankcase on GT3 and TT engines is directly related to the 964 and 993 engines with some minor modifications(the part number on early gt3's starts 964 i beleive). This case has all the bearing surfaces machined into the case which assures concentric running (almost).